BRAFp.V600E, p.V600K, and p.V600R Mutations in Malignant Melanoma: Do They Also Differ in Immunohistochemical Assessment and Clinical Features?

Introduction:Although the detection of BRAF p.V600E mutation by immunohistochemistry was clearly described in melanoma, discordant evidences were reported for the detection of p.V600K and p.V600R mutations. The aim of the study was to evaluate the efficacy of BRAFp.V600E, p.V600K, and p.V600R detection by immunohistochemistry in melanoma. Materials and Methods:Immunohistochemistry with VE1 antibody was performed on 18 tissue samples of metastatic melanomas with known BRAF mutational status. Results:The concordance rate of immunohistochemistry was 100% for p.V600E mutation. In contrast, the 7 p.V600K-mutated melanomas were scored as negative. p.V600K-mutated melanomas were significantly associated with older age, male sex, and worst clinical outcome. Conclusions:Immunohistochemistry could efficaciously be adopted as a first step for the detection of BRAFp.V600E mutation in the initial selection of patients with advanced melanomas as candidates for BRAF inhibitors. It should be followed by molecular techniques in p.V600E-negative melanomas, for the specific search of p.V600K and other non-p.V600E BRAF mutations.

A bout 40% to 60% of malignant melanomas are BRAF mutated. 1,2 The most common mutations are BRAFp.V600E (80%) and BRAFp.V600K (5-30%). 3 These mutations result (mainly p.V600E) in an enhanced BRAF kinase activity and an increased phosphorylation of downstream targets, particularly MEK. Nowadays, the evaluation of somatic BRAF mutations is required for molecular-targeted treatments of metastatic melanoma. In fact, BRAF inhibitors targeting common p.V600E mutations have become increasingly popular because of their high objective response rate and few side effects. BRAF inhibitors have a clinical activity also in melanoma patients harboring a non-p.V600E BRAF mutation, particularly p.V600K and p.V600R. [4][5][6] BRAFp.V600M-D are quite rare and they were not included in the trials for BRAF-selective inhibitors. 5 The ability to recognize BRAFp.V600E mutation changes according to the methods used for mutation testing. At present time, the cobas 4800 BRAFp.V600 Mutation Test, approved by FDA and specifically created to detect BRAFp.V600E mutation, is able to detect 70% of BRAFp.V600K mutations. 7 Although the immunohistochemical assessment of p.V600E mutation was clearly reported in melanoma and other tumors, 8,9 discordant evidences were reported for p.V600K BRAF mutations. 2,10 Different authors reported that none of the non-p.V600E cases, including p.V600K, stained positive with the antibody 8,9 ; in contrast, Routhier et al 10 and Heinzerling et al 11 described 2 p.V600K melanomas with positive VE1 staining.
The aim of this paper was to assess the efficacy of BRAFp.V600E and p.V600K detection by immunohistochemistry in melanoma and the distinctive clinical features of the most common BRAF mutations.

PATIENTS AND METHODS
We selected 35 patients with BRAF-positive advanced melanoma, diagnosed at the University of Modena and Reggio Emilia from January 2010. Among these, 18 patients with known BRAF mutational status (12 males, 6 females) were included ( Table 1). The 22 remaining tissues could not be IHC tested because of nonavailability of enough sample or missing informed consent. We decided to analyze the patients' data and include them in the study because of their epidemiological importance.
Tumor samples were collected according to the principles of the Declaration of Helsinki after obtaining informed consent from each patient. The study was approved by the institutional review board.
Immunohistochemistry with anti-BRAF antibody was performed on 5-mm-thick whole tissue sections of formalin-fixed, paraffin-embedded tissue in a automated immunostainer (Benchmark XT, Ventana) and primary antibodies against BRAFp.V600E (clone: VE1, 1:100) (Spring Bioscience, Pleasanton, CA). All slides were double-blind reviewed by 2 independent observers (A.M.C. and M.M.). The VE1 antibody staining was scored as negative when there was no staining or only isolated nuclear staining. Positive staining was considered by diffuse and moderate (2+) to strong (3+) cytoplasmic staining. Staining results were interpreted as negative when there was no staining (0+), or also slight/faint/ barely perceptible staining or staining of only single cells (1+).

BRAF MUTATION ANALYSIS
Two pathologists reviewed all H&E-stained slides of primary melanomas. The tumor area of interest was detected and marked on each specimen. DNA was extracted using the DNeasy Blood & Tissue Kit (Qiagen, Basel, Switzerland) and tested for BRAF mutations in 2 different laboratories.
Sanger sequencing analyses was performed on all samples to determine BRAF mutational status. The PCR product were checked for the right fragment length, purified and sequenced as previously described. 4 Data were manually edited with the sequencing analysis software (Applied Biosystems, Darmstadt, Germany).

RESULTS
Average ages were 59 years for patients with p.V600E and 66 years for patients with p.V600K (Table 1). BRAFp.V600K-mutated patients had a worse prognosis compared with their V600E counterparts ( Table 2).
The correlation between immunohistochemistry for VE1 antibody and BRAF mutational status is shown in Table 1. V600E-specific antibody stained 6 p.V600Emutated melanomas, but no BRAF wild-type or p.V600K-mutated melanomas. All p.V600E and wildtype cases were correctly identified by the 2 pathologists and the overall concordance rate between protein expression and BRAFp.V600E mutations was 100% (6/6). The single BRAFp.V600R-mutated melanoma had a strong cytoplasmic staining. The 7 BRAF p.V600K-mutated melanomas were reported as negatively stained with VE1 by both observers (Table 1).
In all cases, the interobserver agreement was almost total. Cross reactivity was not observed (Fig. 1).

DISCUSSION
Our study shows that, in contrast to BRAFp.V600E mutation, BRAFp.V600K is not detected by immunohistochemistry with VE1 antibody. Our analysis confirms the high sensitivity and specificity of VE1 antibody in detecting p.V600E mutation, as seen in recently published studies. [8][9][10] Although isolated instances of VE1 reactivity in p.V600K-mutant melanoma have been reported, our findings clarify the value of this antibody as a screening tool for V600E mutations. 10,11 Even though our results are based on a small sample size and further studies are still needed to set appropriate diagnostic standards, we can still highlight some key concepts.
Although BRAF inhibitors are efficacious on tumors with p.V600E and p.V600K mutations, the 2 tumor types shall be considered as distinct entities with slightly different age of onset and clinical behavior. 2,3 Regarding the response to treatment, the clinical trials with vemurafenib/dabrafenib including patients with p.V600K mutations reported worse outcomes and a strong trend for shorter overall survival (OS) among p.V600K melanoma patients compared with those with BRAFp.V600E mutation. 2 Recently published studies demonstrated that patients with BRAFp.V600K mutation are characterized by an older age of cancer onset, an increased risk for brain and lung metastases, and a shorter time from diagnosis to  3 Our case series confirm that p.V600K-mutated melanomas arise at an older age (59 vs. 66 y old) and is predominant in males. In this case, it arises in the head and neck area. Their biological behavior is more aggressive and with tendency to systemic disease, compared with their p.V600E counterparts.
Our experience confirms, according to recent literature on the same topic, the efficacy of VE1 in the recognition of BRAFp.V600R mutation 12 (Table 1, Fig. 1); although just 1 melanoma belonging to our cohort was p.V600R mutated, the clear staining positivity led us to the hypothesis that the protein conformational change is similar enough to the one induced by p.V600E to be bound by VE1 antibody. However, the risk of misidentifying p.V600R as p.V600E by IHC is of limited clinical relevance. p.V600R mutation is less common than the others, but it is the third most common mutation occurring in 5% to 7% of patients with BRAF-mutant melanoma and there are evidences that the melanoma patients carrying this mutation can be successfully treated with oral BRAF inhibitors. 6,12 It is known that it is very useful to screen all patients with advanced melanoma (unresectable stage III and IV) and high risk of recurrence (stage IIIb and IIIc) for BRAFp.V600 mutations other than p.V600E, so that a higher number of patients might benefit from BRAF selective inhibitors. With this regard, BRAFp.V600K mutation, which is present in about 20% of melanomas, should be investigated with a specific antibody at immunohistochemistry or through direct sequencing. In contrast to p.V600E and p.V600R, in fact, VE1 antibody is not able to recognize p.V600K.
To sum up, p.V600K-mutated melanomas seem to constitute a specific clinical and pathologic entity, showing different features in comparison with their p.V600E counterparts. This becomes evident if we consider the different sensitivity to VE1 antibody at immunohistochemistry, which requires the use of distinct antibodies for detecting p.V600K mutation.
Moreover, we realized that the immunohistochemical screening for BRAF mutations, which is less expensive and less time consuming than molecular sequencing, can be efficaciously used in the preliminary screening of the great majority of BRAFp.V600E-mutated melanomas. This preliminary analysis could be then further enriched through the use of a specific antibody for p.V600K mutation and, eventually, direct sequencing for other BRAF mutations in case of negative immunohistochemical staining.